Semantic knowledge affects object memory, but not object-location binding

The recent study demonstrated that object similarity affected object-location memory, but not object memory (Markov & Utochkin, 2021). However, it is not clear to what extent object and object-location memory rely on the independent resources or not. Studies demonstrated mnemonic benefit for familiar objects (Starr et al., 2020). Here we investigated the effects of familiarity on object and object-location memory. We used familiar and unfamiliar objects and their morphs to control for low-level differences. We presented three objects located on the invisible circumference for 1.5 seconds. After a 1.5-second delay, we presented two objects (one previously shown from the set and one new) and asked observers to recognize which was presented. After the recognition task, observers reported the location of the target. We replicated the effect of familiarity on object memory - higher performance for the recognition task was observed for familiar objects. We applied the mixture model with swap (Zhang & Luck, 2008; Bays et al., 2009) and our own MLE model to the data of localization task. We didn’t find any differences between morphed versions of familiar and unfamiliar objects for object and object-location memory, suggesting that our results can’t be explained by low-level differences in stimuli. The number of object-location swap errors was stable across all types of stimuli. Semantic knowledge influences object memory, however the localization errors observed for unfamiliar objects and morphs appear due to “weak” object representation and good guessing strategy (Pratte, 2019)